DTIC ADA549714: Cathodoluminescence Studies of the Inhomogeneities in Sn-doped Ga2O3 Nanowires

0.51 MB·English

by  Defense Technical Information Center

Cathodoluminescence Studies of
the Inhomogeneities in Sn-doped
Ga O Nanowires
2 3
S. I. Maximenko, L. Mazeina, Y. N. Picard, J. A. Freitas, Jr., V. M.
Bermudez and S. M. Prokes
Electronics Science and Technology Division, US Naval Research Laboratory, 4555
Overlook Avenue, Washington, DC 20375, USA
[email protected]
Received Date (will be automatically inserted after manuscript is accepted)
ABSTRACT

Cathodoluminescence real-color imaging and spectroscopy were employed to study the properties of Ga O
2 3
nanowires grown with different Sn/Ga ratios. The structures grown under Sn-rich conditions show large spectral
emission variation, ranging from blue to red, with a green transition zone. Spectral emission changes correlate with
changes in the chemical composition and structure found by energy dispersive spectroscopy and electron
diffraction. A sharp transition from green to red emission correlates with a phase transition of β-Ga O to
2 3
polycrystalline SnO . The origin of the green emission band is discussed based on ab initio calculation results.
2
Recent developments in nanoscience are shifting proven to be a powerful tool for probing the local
towards the formation of doped, core-shell nanostructures characteristics of low-dimensional materials.9-10 The
and heterostructures of semiconductor oxides in order to spatial resolution of CL is defined primarily by the
achieve targeted morphologies, diversity and multi- spherical volume of electron-hole generation created by
functionality.1-8 Non-invasive characterization tools with the electron beam (e-beam) and can be on the order of
high spatial resolution are highly desirable in order to sub-micron in diameter. It was experimentally
study the local structural, optical and electronic properties demonstrated that 20 nm spatial resolution can be
of the given nanoscale systems. The cathodoluminescence achieved in the CL technique.11 In some cases, the CL
(CL) mode of scanning electron microscope (SEM) has sensitivity is four orders of magnitude better than that of
Report Documentation Page Form Approved
OMB No. 0704-0188
Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and
maintaining the data needed, and completing and reviewing the collection of information. Send comments regarding this burden estimate or any other aspect of this collection of information,
including suggestions for reducing this burden, to Washington Headquarters Services, Directorate for Information Operations and Reports, 1215 Jefferson Davis Highway, Suite 1204, Arlington
VA 22202-4302. Respondents should be aware that notwithstanding any other provision of law, no person shall be subject to a penalty for failing to comply with a collection of information if it

does not display a currently valid OMB control number.
1. REPORT DATE 3. DATES COVERED
2009 2. REPORT TYPE 00-00-2009 to 00-00-2009
4. TITLE AND SUBTITLE 5a. CONTRACT NUMBER
Cathodoluminescence Studies of the Inhomogeneities in Sn-doped Ga2O3
5b. GRANT NUMBER
Nanowires
5c. PROGRAM ELEMENT NUMBER
6. AUTHOR(S) 5d. PROJECT NUMBER
5e. TASK NUMBER
5f. WORK UNIT NUMBER
7. PERFORMING ORGANIZATION NAME(S) AND ADDRESS(ES) 8. PERFORMING ORGANIZATION
Naval Research Laboratory,Electronics Science and Technology REPORT NUMBER
Division,4555 Overlook Avenue SW,Washington,DC,20375
9. SPONSORING/MONITORING AGENCY NAME(S) AND ADDRESS(ES) 10. SPONSOR/MONITOR’S ACRONYM(S)
11. SPONSOR/MONITOR’S REPORT
NUMBER(S)
12. DISTRIBUTION/AVAILABILITY STATEMENT
Approved for public release; distribution unlimited
13. SUPPLEMENTARY NOTES
Nano Letters 9 (2009) 3245
14. ABSTRACT

15. SUBJECT TERMS
16. SECURITY CLASSIFICATION OF: 17. LIMITATION OF 18. NUMBER 19a. NAME OF
ABSTRACT OF PAGES RESPONSIBLE PERSON
a. REPORT b. ABSTRACT c. THIS PAGE Same as 7
unclassified unclassified unclassified Report (SAR)
Standard Form 298 (Rev. 8-98)
Prescribed by ANSI Std Z39-18
x-ray microanalysis techniques, and impurities with
concentrations around 1014·cm-3 or lower can be
detected.12 This makes CL a valuable non-destructive tool
for studies of inhomogeneities in nanostructures
introduced by doping and/or defects. In most cases, CL
characterization is limited to simple panchromatic
imaging, which does not provide clear information about
spectral luminescence variation across the studied
specimen. Monochromatic CL imaging and spectroscopy
are often implemented in order to obtain this important
information. However, monochromatic CL imaging is
time consuming and requires multiple scans to obtain
comprehensive data that may introduce additional defects
due to excessive e-beam radiation.
In this work, real-color CL imaging (RC-CL) and CL
spectroscopy were employed to study the properties of

undoped (unintentionally doped) β-Ga O , Sn-doped β-
2 3
Ga O and Ga O -SnO nanowires (NWs). Sn-doped β-
2 3 2 3 2
Figure 1. RC-CL image of undoped β-GaO nanowires.
Ga O NWs showed uniform luminescence which was 2 3
2 3
red-shifted by 30 nm relative to that of undoped β-Ga O
2 3
the Ga O -SnO heterostructures (HS) was performed
NWs. The spectral luminescent variation along the Ga O - 2 3 2
2 3
by SIMS and EDS. Structural characterization was
SnO NS was correlated with local structural and
2
conducted using a Oxford Instruments Nordlys II EBSD
chemical properties studied by other SEM-based
system using a 20 kV, 2.4 nA e-beam. The influence of Sn
techniques such as electron backscatter diffraction
on the optical properties of Ga O was verified by ab
(EBSD) and energy dispersive spectroscopy (EDS). It is 2 3
initio theoretical calculation performed using the
demonstrated that the combination of RC-CL and CL
procedure described elsewhere.19
spectroscopy provides a powerful, sensitive, and non-
The room-temperature (RT) RC-CL of as-grown
destructive approach to identify inhomogeneities and to
undoped monoclinic β-Ga O NWs (the crystalline
study intrinsic and extrinsic properties of nanostructured 2 3
structure identified by TEM, EBSD and FTIR analyses15)
materials.
showed uniform blue emission along the NW lengths (Fig.
The structures studied in the present work were grown
1). The emission consisted of a broad asymmetric
at 900 °C by the vapor-liquid-solid (VLS) mechanism
under Ar atmosphere as described elsewhere13 using Gaussian-shaped band with a maximum around 420 nm
silicon (100) wafers covered with 20 nm gold films
serving as substrates. Undoped β-Ga O nanowires were
2 3
grown using Ga (99.995%)14,15 while Sn doped β-Ga O
2 3
NWs were grown using 98:2 wt% Ga-Sn alloy (99.999%,
Alfa Aesar) as sources.13 Synthesis of the Ga O –SnO
2 3 2
nanowire heterostructures was performed under the same
conditions as described above but using metallic Sn
(99.999%, Alfa Aesar) and pure Ga (99.9999%) placed in
different zones of the furnace in order to increase the
Sn/Ga ratio in the vapor phase.13
A conventional LEO 435VP thermal emission SEM
equipped with a retractable parabolic collecting mirror
was used for the luminescence measurements. A constant-
flow liquid helium cold stage was employed to vary the
sample temperature in the range between 5 K and 330 K.
A probe e-beam with 10 kV accelerating voltage and 1 nA
current was used to excite carriers in the target samples
for optimal CL signal-to-noise ratio and high spatial
resolution. The latter was determined by the size of the
electron-hole pair generation sphere and was estimated for
our experimental conditions to be ~0.4 µm using the
Kanaya and Okayama model.16 CL imaging was acquired
by piping the light emitted by the sample to three
fundamental color channels, calibrated to compose real-
color CL imaging.17 The CL spectra were obtained by

redirecting the emitted light to a Triax550 spectrometer
fitted with a liquid nitrogen cooled CCD camera.18
Figure 2. CL spectra of individual undoped and Sn-doped β-
The chemical analysis of the Sn-doped Ga O NWs and
2 3 GaO NWs measured at (a) 293 K and (b) 5K.
2 3
band was observed at 700 nm (1.77 eV). This band is
easily observed in the low temperature CL spectrum (Fig.
2), due to the reduction of the linewidth of emission
bands. This band may be related to DAP recombination
processes involving electrons trapped by oxygen
vacancies and holes captured on deep acceptor levels
introduced by nitrogen impurities.25
Sn-doped Ga O NWs, grown using the GaSn alloy,
2 3
were formed by the VLS process, where the Sn and Ga
from the vapor formed a liquid eutectic droplet with the
Au catalyst, leading to a relatively strain and distortion
free growth of the monoclinic β-Ga O nanowire
2 3
structure, as confirmed by TEM, EBSD and FTIR
analyses.13 The NWs exhibit morphologies similar to that
of pure β-Ga O (straight with diameters 100-250 nm and
2 3
lengths of several microns). This NWs have trace amounts
of Sn (0.2 wt%) revealed by chemical analysis of the
material. The CL spectra depicted in Fig. 2 showed that
the emission peak is red shifted to 450 nm and 390 nm for
room and low temperatures, respectively. Similar to
undoped Ga O NWs (Fig. 1), no emission spectra
2 3
variations along the length of single NWs or throughout
the sample were detected indicating that no Sn
segregation occurred. The observed red shift in the
emission upon Sn doping correlates with previously
reported results, where the introduction of Sn impurity in
β-Ga O powder promoted a shift of the peak of the
2 3
emission band from blue to green.26 Also, it was
suggested that the red shift can be attributed to a number
of oxygen vacancies presented in the material.9,27 Despite
numerous experimental studies, the recombination
Figure 3. (a) SEM and (b) RC-CL images of Sn-doped GaO
2 3 mechanism responsible for the green emission is still not
NWs grown using the GaSn alloy. Note the color variation
along individual wire. understood. Possible mechanisms of the green emission

(2.95 eV) and an exponential tail at the lower energy side
(Fig. 2). This band becomes narrower and shifts to 380
nm (3.26 eV) at low temperature (~5K). Note that these
values are well below the reported near-edge
recombination energy value for wide band gap β-Ga O
2 3
(Eg=4.9 eV).20 The broad blue emission band is a
distinctive signature of intrinsic n-type conductivity for
bulk and nanocrystaline monoclinic β-Ga O 21,22 and was
2 3
assigned to recombination processes between donor-
acceptor pair (DAP) centers, where donor and acceptor
levels are introduced by oxygen and gallium vacancies,
respectively.21 Characteristic n-type conductivity found in
undoped β-Ga O has been associated with electrons
2 3
trapped in oxygen vacancies.23 The absence of the
emission lines at higher energies in the spectra provides
evidence that interband emission due to recombination of
electron-hole pairs (near bandedge emission) is
suppressed by the presence of higher concentration of
compensating deep levels. The DAP band is intrinsically
broad due to variation of the energy associated with the
Columbic interaction between individual donors and
acceptors involved in the recombination process.
Additional line broadening can be introduced by the large
spatial fluctuation in the band edges, due to statistical
fluctuation in the local internal fields produced by Figure 4. Characteristic CL spectra of blue, green and red
fluctuations in the density of charged impurities.24 In emission regions of the nanowires measured at (a) room
temperature and (b) at 5K.
addition to the blue band, a low-intensity red emission
Figure 5. (a) SEM, (b) RC-CLi and EDS analysis of distribution of O (c), Ga (d), Sn (e) and Au (f) acquired from the same nanowire.
Arrows on (d) and (e) indicate an increase in the concentration. EDS spectra captured from (g) blue, (h) green and (i) red emission parts of
the wire as indicated by arrows in (b). Si signals are generated from the substrate.

will be discussed later in this paper. To identify the mechanisms associated with the large
RC-CL of NWs grown under Sn-rich conditions variations in the CL emission spectral distribution, a
revealed that NWs have large spectral emission number of analyses probing the chemical composition and
variations, covering the spectral range from blue to green crystalline structure of different parts of NS were carried
to red. Note that greater color variations and out. Analysis of the chemical composition by EDS
thickness/shape inhomogeneities are especially revealed only O, Ga, Sn and Au elements in the wires.
pronounced in NWs with larger diameters (500-600 nm) Characteristic distribution of the elements and their
(Fig. 3b) while small diameter NWs (20-100 nm) showed correlation with RC-CL for one of these nanowires is
more uniform luminiscence in both straight and kinky presented in Fig. 5. The NW region characterized by
parts. High resolution RC-CL of the wires (Fig. 3b, insert) intense blue emission is rich in Ga with trace amounts of
revealed that while the transition from blue to the green Sn, with Sn content gradually increasing toward the green
emission is gradual, the green to red transition has a region to ~1.5-3 wt% (Fig. 5b, d, e). The shift of the
noticeable sharp boundary. The red emission is clearly spectral peak emission position to lower energy with
dominant at the top region of the wires, while the regions increased doping concentration is typically associated
closer to the Si substrate show increasing blue emission. with bandgap reduction due to heavy doping.12 The Sn
Room and low temperature CL spectroscopy data concentration is highest at the top of the NWs, where the
acquired in the regions with each individual color red emission dominates (Fig. 5d), while the density of Ga
emission (blue, green, red) are depicted in Fig. 4. Low atoms is significantly decreased (down to ~5 wt%). No
temperature CL spectra (Fig. 4b) measured in the blue significant variations in the oxygen content were found
region of a single NW showed an emission band with a along the NW (Fig. 5c). The characteristic signature
peak around 394 nm (3.15 eV), which corresponds to the indicating the VLS growth mechanism is the Au cap
peak observed in spectra of Sn doped β-Ga O NWs (Fig. observed at the tips of the NWs (Fig. 5f).
2 3
3). The CL spectrum of the “green” region (middle part of The EBSD characterization of the Ga O -SnO NWs
2 3 2
the NW) has a peak at around 500 nm (2.48 eV). The CL revealed that the parts of the wires with the dominant blue
spectra acquired at the center of the “red”-region has a emission have the β-Ga O crystalline structure (Fig. 6a).
2 3
maximum intensity near 600 nm (2.07 eV). The transition part of the wires with the green emission,
which had a high Sn content, was also found to have a β-
Figure 6. EBSD patterns recorded from (a) blue (b) green and (c) red regions of the GaO-SnO nanowires.
2 3 2

Ga O crystalline structure (Fig. 6b). However, Kikuchi time, the relative Sn concentration increases in the vapor
2 3
bands in the recorded EBSD patterns were noticeably phase. Thus, higher amounts of Sn begin to incorporate
weaker, indicating significantly poorer crystalline quality into the already formed β-Ga O structure, inducing strain
2 3
in the green region. The red emission region was and distortions that will produce lower crystalline quality
identified as rutile-type SnO with numerous variations in β-Ga O . As more Sn continues to dissolve into the Au
2 2 3
orientation, indicating that the SnO was polycrystalline catalyst, a polycrystalline SnO phase begins to form
2 2
(Fig. 6c). The emission band with peak at ~600 nm (2.07 which also contains some Ga. Thus, each Ga O -SnO
2 3 2
eV) is consistent with previously reported spectra data for nanowire consists of three distinct parts: monocrystalline
SnO .28 The observed luminescence is probably related to Sn-doped β-Ga O , poorly crystalline Sn-doped β-Ga O ,
2 2 3 2 3
DAP recombination between electrons trapped by oxygen and polycrystalline Ga-doped SnO . The formation of
2
vacancies and holes captured on deep acceptor levels single ε-Ga O NWs during this particular growth process
2 3
introduced by the Ga impurities. The broad emission band is believed to be related to smaller Au catalyst particles
may result from the polycrystalline nature of the wire. It nucleating smaller diameter NWs with larger surface
should be mentioned that TEM analysis of the blue- areas, thermodynamically favoring ε-Ga O formation
2 3
emitting straight parts of the small diameter wires over β-Ga O .13 Additional studies of these ε-Ga O NWs
2 3 2 3
indicated that these regions crystallized as orthorhombic are underway to probe variations in Sn doping
ε-Ga O .13 Though no experimental data are available for concentration.
2 3
the band gap value of ε-Ga O , computational studies Rutile tin oxide is an intrinsic n-type direct wide-band-
2 3
showed that this value is close to that of β-Ga O .29 The gap (Eg=3.6 eV) semiconductor oxide.30 The electrical
2 3
conductivity of tin oxide results primarily from the
observed blue luminescence band from ε-Ga O , which
2 3
existence of oxygen vacancies (V ), which act as
correlates with the emission from the low Sn-doped β- O
donors.31 However, Ga3+ or Ga2+ are substitutional
Ga O sample, probably has the same origin as that of β-
2 3 impurities, introducing acceptor energy levels within the
Ga O .
2 3 forbidden gap of SnO and therefore increasing hole
Based on the obtained CL, EBSD and EDS results, we 2
concentration.31 Thus, a transition from n- to p-type
conclude that the factors responsible for the self-assembly
conductivity is possible for SnO with an increasing
of the Ga O -SnO structures are the different vapor 2
2 3 2 concentration of Ga This could lead to the formation of n-
pressures of Sn and Ga as well as different growth .
p heterojunction in a single Ga O -SnO nanowire.
kinetics for Ga O and SnO . First, Ga dissolves in the Au 2 3 2
2 3 2 Based on our observations and previous literature
catalyst, resulting in the growth of a well-ordered Sn-
reports,26 Sn incorporation causes a red shift of the blue
doped β-Ga O NW. Due to the Ga consumption in the
2 3
growth, the Ga concentration decreases and at the same
an unstable electronic structure. No such problems were
encountered for substitution at a tetrahedral site. The Sn+3
involves a single unpaired electron; hence, all calculations
were spin-unrestricted. The first step was to relax the
geometry to a minimum in the total energy in a Hartree-
Fock calculation in which the unit-cell dimensions were
fixed but all atoms were free to move. For Ga and O,
Durand-Barthelat effective-core pseudopotentials (ECPs)
were used with the basis sets optimized for β-Ga O .19 Sn
2 3
was described using the Hay-Wadt large-core ECP with
the basis set optimized for SnO .33 After geometry
2
optimization, a single-point density functional theory
(DFT) calculation was done using the B3LYP hybrid
functional and the all-electron Ga and O basis sets given
in Ref. 19 together with the same ECP for Sn. The density
of states was then computed using the DFT wavefunction.
The calculation results revealed a Sn-derived state at
the Fermi level which lies at about 0.2E below the
g
conduction band minimum and also a Sn contribution
throughout the valence band as presented in Fig. 7. The
fully-relaxed electronic configuration of the Sn impurity
resembles that of a Sn+2 ion. The CL emission can then be
interpreted in terms of the recombination of an electron in
the Sn-induced gap state with a hole at the top of the
valence band. The emission energy of ∼3.8eV estimated
from this model (using the experimental E value of
g
∼4.9 eV) is an upper limit since it neglects any excitonic
effect. Although we have not exhaustively studied all
possible Sn-related defects, nor analyzed the defect
formation energies, the results demonstrate that sub-band-
gap emission can reasonably be ascribed to the presence
Figure 7. Computed (a) total and (b) Sn projected density of of a Sn impurity. The gradual shift from blue to green
states. The Sn plot shows the virtually identical spin-up and emission is probably a result of a competition between
spin-down components. The zero energy is at the Fermi level,
intrinsic recombination process involving DAP (donor
and the calculated band gap is Eg= 5.8eV.
and acceptor levels introduced by the oxygen and the

gallium vacancies) and extrinsic transition from Sn donor
level to free holes.
emission band in Ga O . It was suggested that the green
2 3 In Figure 8 we present schematically the proposed
emission band may be related to DAP recombination
assignments of various radiative transitions observed in
processes, where substituting Ga+3 by multivalent Sn
the present work for β-Ga O and SnO . The
results in divalent and tetravalent valence states of Sn 2 3 2
recombination channel (a), which is the origin of blue
which act as donors and as acceptors.26 However the lack
emission, arises from transitions in β-Ga O between
of complete supporting experimental evidence and 2 3
electron trapped by oxygen vacancies (donor level) and
theoretical studies focused on modeling the energetics of
Sn impurity in Ga O leave this topic as an open question.
2 3

In order to verify that a Sn impurity in Ga O could be the
2 3
source of the red-shift in the CL emission band,we
performed ab initio theoretical calculations. Only the
simplest type of defect was considered; namely, a single
Sn+3 substituting for Ga+3. The model consisted of a
periodic lattice built on a supercell of dimensions 1a x 4b
x 2c, where a = 12.202 Å, b = 3.035 Å and c = 5.799 Å
are the crystallographic unit-cell dimensions of β-Ga O .19
2 3
The supercell contained 16 Ga O units, giving a Sn/Ga
2 3
ratio of 1/31 or ∼3% doping. Previous simulation studies
of point defects in Ga2O3 using a shell model32 revealed
that in monoclinic β-Ga O , where Ga atoms occupy both
2 3
octahedral and tetrahedral sites, Sn+4 ions prefer to Figure 8. Schematic diagram with the assignments of radiative
substitute Ga in octahedral sites. Therefore in our transitions observed in the studied samples (see text for details).
calculations the Sn+3 was initially substituted at an
octahedral Ga+3 site, but this led to severe problems in the
convergence of the self-consistent field (SCF) indicating
holes captured by gallium vacancies (acceptor level). (13) Mazeina, L.; Picard, Y. N.; Maximenko, S. I.; Perkins, F. K.,
The red emission, associated with the recombination Glaser, E. R., Twigg, M. E; Freitas, J. A; Prokes, S. M. Submitted:
Crystal Growth and Design.
channel (b), results from transitions between electrons (14) Prokes, S. M.; Carlos, W. E.; Glembocki, O. J. Proc. SPIE Intern.
trapped by theoretical studies focused on modeling the Soc. Opt. Enginer. 2005, 6008, 60080C/1.
energetics of oxygen vacancies and holes captured on (15) Mazeina, L.; Picard, Y. N.; Prokes, S. M. Cryst. Growth Design
deep acceptor levels introduced by nitrogen impurities. 2009, 9, 1164.
(16) Kanaya, K.; Okayama, S.; J. Phys. D: Appl. Phys. 1972, 5, 43.
The recombination channel (c) is related to transitions
(17) Saparin, G. V.; Obyden, S. K. Scanning 1988, 10, 87.
between donor level introduced by the incorporation of Sn (18) Maximenko, S. I.; Freitas Jr., J. A.; Klein, P. B.; Shrivastava, A.;
atoms in Ga tetrahedral sites of β-Ga O and holes at the Sudarshan, T. S. Appl. Phys. Lett. 2009, 94, 092101.
top of the valence band. The red em2iss3ion in SnO may (19) Bermudez, V.M. Chem. Phys. 2006, 323, 193.
2 (20) Matsumoto, T.; Aoki, M.; Kinoshita, A.; Aono, T. Jpn. J. Appl.
result from recombination between the electrons trapped
Phys. 1974, 13, 1578.
by the oxygen vacancies and holes captured on the deep (21) Binet, L.; Gourier, D. J. Phys. Chem. Solids 1998, 59, 1241.
acceptor levels introduced by Ga impurities, as indicated (22) Nogales, E.; Mendez B.; Piqueras, J. Appl. Phys. Lett. 2005, 86,
by recombination channel (d) in Fig. 8. 113112.
(23) Tippins, H.H. Phys. Rev. 1965, 140, A316.
In summary, we demonstrated the capabilities of SEM-
(24) Dean, P.J.; Proc. in Sol. State Chem. 1973, 8, 1126
based RC-CL imaging, in combination with point CL (25) Song, Y. P.; Zhang, H. Z.; Lin, C.; Zhu, Y. W.; Li, G. H.; Yang, F.
spectroscopy, for the detection and identification of local H.; and Yu, D. P. Phys. Rev. B 2004, 69, 075304.
changes of nanostructured material properties. Sn-doped (26) Harwing, T.; Kellendonk, F. J. Solid State Chem. 1978, 24, 255.
(27) Villora, E.G.; Atou T.; Sekiguchi, T.; Sugawara, T.; Kikuchi, M.;
Ga O NWs showed uniform luminescence that is 30 nm
2 3 Fukuda, T.; Solid State Commun. 2001, 120, 455.
red-shifted relative to undoped Ga2O3 NWs. Ga2O3-SnO2 (28) De Murcia, M.; Egeet, M.; Fillard, J. P. J. Phys. Chem. Solid.
nanowire heterostructures are characterized by large 1978, 39, 629.
optical and structural property variations along the length (29) Yoshioka, S.; Hayashi, H.; Kuwabara, A.; Oba, F.; Matsunaga,
K.;Tanaka, I. J. Phys.: Condens. Matter. 2007, 19, 346211.
of each individual wire. Three distinct regions
(30) Batzill, M.; Diebold, U. Prog. Surf. Sci. 2005, 79, 47.
characterized by blue, green and red emissions with peak (31) Bagheri-Mohagheghi, M. –M.; Shokooh-Saremi, M. In Trends in
at 390 (3.18 eV), 500 (2.48 eV) and 600 nm (2.07 eV), Semiconductor Research; Elliot, T. B., Ed.; Nova: Hauppauge
respectively, were observed. A sharp transition region (NY), 2005.
with emission bands from green to red was observed. EDS (32) Blanco, M. A.; Sahariah, M. B.; Jiang, H.; Costales, A.; Pandey,
R.; Phys. Rev. B 2005, 72, 184103.
and EBSD analyses indicated that these bands are (33) Bredow, T.; Pacchioni, G. Theor. Chem. Acc. 2005, 114, 52.
associated with phase transition from Sn-doped
monoclinic Ga O to Ga-doped rutile SnO . Ab initio
2 3 2
calculations demonstrated that the Sn impurity can
introduce an energy level in the forbidden gap of β-Ga O
2 3
lying ~0.2 E below the conduction band minimum,
g
leading to a donor-to-free-hole transition which is most
likely responsible for the spectral red shift of emission
band in Sn-doped β-Ga O NWs.
2 3
Acknowledgment S. I. Maximenko, L. Mazeina and Y.
Picard thank the National Research Council (NRC)
program for their administrative support. This work was
partially supported by the Office of Naval Research
(ONR).

(1) Fan, H. J.; Fleischer, F.; Lee, W.; Nielsch, K.; Scholz, R.;
Zacharias, M.; Goesele, U.; Dadgar, A.; Krost, A. Superlatt.
Microstr. 2004, 36, 95.
(2) Lao, J. Y.; Wen, J. G.; Ren, Z. F. Nano Lett. 2002, 2, 1287.
(3) Gao, P.; Wang, Z. L. J. Phys. Chem. B 2002, 106, 12653.
(4) Kuan, C. Y.; Chou, J. M.; Leu, I. C.; Hon, M. H. J. Sol. St. Chem.
2008, 181, 673.
(5) Zhong, L. W. Mater. Today 2004, 6, 26.
(6) Xu, L.; Su, Y.; Li, S.; Chen, Y.; Zhou, Q.; Yin, S.; Feng, Yi. J.
Phys. Chem. B 2007, 111, 760.
(7) Zhang, D.-F.; Sun, L.-D.; Jia, C.-J.; Yan, Z.-G.; You, L.-P.; Yan,
C.-H. J. Amer. Chem. Soc. 2005, 127, 13492.
(8) Lugstein, A.; Andrews, A. M.; Steinmair, M.; Hyun, Y.-J.;
Bertagnolli, E.; Weil, M.; Pongratz, P.; Schramboeck, M.; Roch,
T.; Strasser, G. Nanotechn. 2007, 18, 355306/1.
(9) Yu, D. P.; Bubendorff, J.-L.; Zhou, J. F.; Leprince-Wang, Y.;
Troyon, M. Solid State Commun. 2002, 124, 417.
(10) Dovrat, M.; Arad, N.; Zhang, X.-H.; Lee S.-T.; Sa'ar, A. Phys.
Rev. B 2007, 75, 205343.
(11) Norman, C. E. Solid State Phenomena 2001, 78-79, 19.
(12) Yacobi, B. G.; Holt, D. B. Cathodoluminescence Microscopy of
Inorganic Solids; Springer: Berlin, 1990.